Impact of different models of improvement of continuity of care on lipid control and the delay of visits to cardiology

Due to the ageing of the population and the better treatment of acute conditions, chronic diseases are dramatically increasing worldwide. As a result, it is necessary a comprehensive approach in order to assure the sustainability of the healthcare system [1]. Cardiovascular disease is associated with significant morbidity and disability and it is the leading cause of mortality [2]. In addition, despite current treatments, rehospitalizations after an acute cardiovascular event are frequent, leading to an impairment of the quality of life of patients and an increase of healthcare costs [3,4].

Assuring and adequate continuity of care for people experiencing chronic conditions is mandatory to provide the best management of this population, as this may improve the quality of care and clinical outcomes [5–7]. Unfortunately, traditional continuity of care has many gaps and integration between different healthcare levels is far from optimal [8–11]. This insufficient coordination between primary care and other specialties leads to a heterogeneity in the management of patients with chronic conditions, unsuccessful achievement of therapeutic goals, duplication of complementary tests, unnecessary visits to the specialist, dissatisfaction of patients with medical attention, and importantly, to more adverse outcomes [10,11]. Although different initiatives to improve healthcare integration have been developed, it remains unknown which continuity of care model may improve this integration to a greater extent [12–16].

Current European guidelines recommend reducing LDL cholesterol levels to less than 70 mg/dl among patients at very high risk of cardiovascular events, such as those with ischemic heart disease or atherosclerotic stroke [2]. Unfortunately, the majority of patients with atherosclerotic cardiovascular disease do not achieve LDL cholesterol goals [17–19]. It is likely that a better coordination between healthcare levels could improve this insufficient control of cardiovascular risk factors among patients with atherosclerotic vascular disease.

The aim of this study was to analyze the impact of implementing three different models of continuity of care in three hospitals on the delay of first visits to the cardiologist (management end point) and on LDL cholesterol control rates among patients with atherosclerotic vascular disease (clinical end point).

Methods

For this purpose, a retrospective analysis of the implementation of three different models of continuity of care in three hospitals (Arnau de Vilanova Hospital, Valencia, Moises Broggi Sant Joan Despi Hospital, Barcelona and Álvaro Cunqueiro Hospital, Vigo) was performed. A total of 300 patients with atherosclerotic cardiovascular disease were included in the study (111 from the Arnau de Vilanova Hospital, 116 from the Moises Broggi Sant Joan Despi Hospital, and 74 from the lvaro Cunqueiro Hospital). In the Arnau de Vilanova Hospital and Álvaro Cunqueiro Hospital, a virtual system (telecardiology) was developed. The virtual system consisted of telephone calls and/or website requests in which the cardiologist resolved the questions that primary care physicians may have, and when this was not possible, the patient was referred to the cardiologist. In addition, in the Arnau de Vilanova Hospital, internal audits and specific medical education programs for improving lipid control in primary care were performed. In the Moises Broggi Sant Joan Despi Hospital, a staff cardiologist was integrated into the primary care center and attended patients every week in the corresponding primary care center (on-site and virtual visits). In addition, discussion of clinical cases as well as ongoing medical education activities were also performed by each cardiologist in the primary care center. In the three hospitals, the different models of continuity of care were implemented at the same time during 2013. In this study, the impact of the three models on continuity of care, the delay of visits to the cardiologist and LDL cholesterol control rates among patients with atherosclerotic cardiovascular disease were analyzed before the implementation (2013) and after 2 years of the implementation of the programs (2015). This study was performed according to the principles outlined in the Declaration of Helsinki.

To determine the impact of the programs on the visits to cardiology (management end point), the mean delay of first visits to the cardiologist in the year before the implementation (2013) and 2 years after the implementation of the programs (2015) were determined.

To analyze the impact of the programs on lipid control (clinical end point), patients with atherosclerotic cardiovascular disease and LDL cholesterol ≥70 mg/dl were randomly and retrospectively selected. LDL cholesterol levels before the implementation of the programs (2013) and 2 years after the implementation of the continuity of care models (2015) were compared. The proportion of patients that achieved LDL cholesterol goals (<70 mg/dl) was determined at study end. In addition, baseline clinical characteristics of the patients (age, sex, hypertension, diabetes and smoking) were recorded. The time (<1 year vs >1 year) in which patients had had the atherosclerotic cardiovascular event (acute coronary syndrome, angina, stroke) was also determined. All data were recorded in the overall study population and compared between the three hospitals.

Statistical analysis

Categorical variables were expressed as absolute numbers and percentages and quantitative variables as mean and standard deviation. To compare categorical variables, the χ² test or the Fisher test were used, according to the sample size. To compare continuous variables, the Student t test or the Mann–Whitney U test were performed, as appropriate. A two-tailed p-value <0.05 was considered to indicate statistical significance. The Bonferroni's method was performed to adjust for chance effects. The statistical analysis was performed using the SPSS statistics package, version 23.0 (SPSS, IL, USA).

Results

Overall, the delay of the visits to the cardiologist improved from 66.5 ± 29.1 days before the implementation of the continuity of care models, to 34.1 ± 14.1 days after the implementation (relative reduction of 48.7%; p < 0.001). Although there was a significant reduction of the delay of the visits to the cardiologist in the three hospitals, this reduction was significantly higher in the Moises Broggi Sant Joan Despi Hospital (54.5%; p < 0.001 vs baseline) compared with the Arnau de Vilanova Hospital (38.8%; p < 0.001 vs baseline) and Álvaro Cunqueiro Hospital (46.4%; p = 0.002 vs baseline; Table 1).

A total of 300 patients with atherosclerotic cardiovascular disease were included in the study (111 from the Arnau de Vilanova, 116 from the Moises Broggi Sant Joan Despi Hospital, and 74 from the Álvaro Cunqueiro Hospital). Overall, mean age was 64.0 ± 7.5 years, 76.0% were men, 82.3% had hypertension, 34.3% diabetes and 15.7% were current smokers. The majority of patients (86.1%) had had the cardiovascular event more than one year before being included in the study. No significant differences were observed regarding age, sex or the proportion of patients with diabetes between the three hospitals, but hypertension was more common in the Moises Broggi Sant Joan Despi Hospital, current smokers in the Arnau de Vilanova Hospital and the percentage of patients with the cardiovascular event <1 year before inclusion in the Álvaro Cunqueiro Hospital (Table 2).

Lipid lowering drugs before and after the implementation of the programs were reported in Figure 1. Overall, before the implementation, 47.0% of patients were taking intermediate-intensity statins, 31.4% high-intensity statins, 7.1% combined therapy and 7.1% of patients were not taking any lipid lowering drug. After the implementation of the programs, these numbers significantly moved to 29.8, 42.4, 22.4 and 2.4%, respectively. The proportion of patients taking high-intensity statins or combined therapy increased from 38.5% before the implementation of the programs to 64.8% (p < 0.001) and this increase occurred in both, patients with the cardiovascular event <1 year before inclusion (52.4 vs 71.4%; p = 0.07) and >1 year before inclusion (35.7 vs 62.4%; p < 0.001). Although there were some differences in the prescription of lipid lowering drugs at baseline between hospitals, after the implementation of the programs, the prescription of lipid lowering drugs was similar between hospitals (Table 3).

Overall, mean LDL cholesterol levels decreased from 99.9 ± 27.3 mg/dl before the implementation of the programs to 80.4 ± 22.7 mg/dl after the implementation of the programs (absolute reduction of -19.5 ± 28.6 mg/dl; relative reduction of 19.5%; p < 0.001). At baseline, LDL cholesterol levels were similar between groups. Although in the three hospitals there was a significant reduction of LDL cholesterol levels after the implementation of the programs, this reduction was higher in the Arnau de Vilanova Hospital (p = 0.004; Table 4).

At baseline, all patients had an LDL cholesterol ≥70 mg/dl. After the implementation of the programs, 35.3% of patients reached LDL cholesterol goals. The proportion of patients that achieved LDL cholesterol targets was higher in the Arnau de Vilanova Hospital (46.8%) compared with the other hospitals (26.1% in the Moises Broggi Sant Joan Despi Hospital and 32.4% in the Álvaro Cunqueiro Hospital; p = 0.004). Overall, there was a trend toward higher LDL cholesterol control rates among patients that had had the cardiovascular event <1 year before being included in the study compared with those that had had the cardiovascular event >1 year, but did not reach statistical significance (Table 5).

Discussion

Overall, our study showed that enhancing continuity of care between cardiology and primary care led to improvements of management and clinical end points. Thus, the delay of the first visits to the cardiologist decreased approximately by 49% after only 2 years of implementing the programs. In a recent study, the integrated care model was associated with a reduction of the delay not only of the first visits but also the follow-up visits to the cardiologist. This reduction of the delay of visits allowed an earlier treatment of patients [20].

Integration models improve the management of patients with chronic cardiovascular conditions, including an increase of cardiovascular risk factors control rates, and a higher use of guideline-recommended therapies, but importantly, without an increase in the use of resources [14,21,22]. However, changes in aspects of employment, organization and improving interaction and communication between physicians are required in order to assure a more efficient use of available resources [23,24]. In addition, those centers with a higher level of integration between primary care and cardiology benefit more from a greater communication and satisfaction [25].

In our study, three different continuity of care models were compared (virtual visits [telecardiology], virtual visits together with internal audits and specific medical education programs and integration of the cardiologist into the primary care center). Although the three models were associated with a significant reduction of the delay of fist visits to cardiology, it seems that integrating the cardiologist into the primary care center would be more effective than telecardiology. Improving communication with primary care should be considered as a target by itself, as it is associated with better outcomes [14–16,26]. This communication between healthcare levels should include not only intensive educational sessions, but also new technologies [27], such as telecardiology, telemonitoring of patients or even cardiac telerehabilitation, and as our study showed, the integration of at least one cardiologist into the primary care center [20,28,29]. However, to assure a better continuity of care, all these conditions should be adapted to each specific sanitary area and healthcare system. The continuity of care between healthcare levels should be periodically evaluated to determine gaps and improvement areas. Thus, specific questionnaires have been developed for the measurement of continuity of care between healthcare levels [30,31].

A recent study that assessed the proportion of patients with atherosclerotic cardiovascular disease who would require intensification of lipid lowering therapies for achieving LDL cholesterol goals showed that with current treatments, the great majority of patients should attain recommended therapeutic targets [32]. Different studies have shown that in routine practice, less than 30% of patients with ischemic heart disease meet the LDL cholesterol treatment target of less than 70 mg/dl [17,18,33–35]. This is not limited to patients with ischemic heart disease, but also occurs in patients with atherosclerotic disease in other vascular beds [19]. Remarkably, despite the current concern about the importance of achieving lipid goals in high-risk patients and the potent lipid lowering drugs that are available at this moment, the control of cardiovascular risk in coronary artery disease patients has only slightly improved in the last decades [36]. This is mainly because of the low use of high dose statins or combined lipid-lowering therapy in routine practice. In fact, this has also been reported in the EUROASPIRE V, the last large registry of patients with ischemic heart disease in Europe [35]. In our study, the implementation of the new continuity of care models translated into an intensification of lipid lowering therapy (the proportion of patients taking high-intensity statins or combined therapy significantly increased from 38.5 to 64.8%), and this made that 35% of patients reached LDL cholesterol goals. Although this could be considered a low percentage, it is important to emphasize that, in contrast to previous studies, at baseline, before the implementation of the programs, all patients had an LDL cholesterol ≥70 mg/dl (mean LDL cholesterol 99.9 mg/dl).

Importantly, there was a trend toward higher LDL cholesterol control rates among patients that had had a recent cardiovascular event. Despite patients with atherosclerotic cardiovascular disease have a very high risk of new cardiovascular events, the fact is that underestimation of risk is frequent in this population, and this is more common as time passes from the acute event [37,38], leading to a worse lipid control, as our data suggested. In fact, it has been reported that although in the majority of patients with an acute myocardial infarction, statin therapy is prescribed, rates of statin intensification and maximization are low, particularly among patients with chronic ischemic heart disease, and this is associated with poorer outcomes [39–41]. Although therapeutic inertia is common in clinical practice [17], improving coordination between primary care and cardiology promotes the intensification of treatment and the achievement of lipid control rates, as our study showed.

Despite the implementation of the three new models of continuity of care were associated with significant reductions of LDL cholesterol values, this reduction was significantly greater when telecardiology together with internal audits and specific medical education programs were developed (vs telecardiology alone or integration of the cardiologist into the primary care center). This means that improving communication between healthcare levels, although necessary [10,11], it is insufficient to achieve therapeutic goals, and continuous ongoing medical education is mandatory. In addition, it is likely that internal audits make physicians to realize about the importance of achieving LDL cholesterol goals and the need for intensifying lipid lowering therapy. However, considering that current control rates are still far from optimal, more efforts are needed to improve these numbers.

This study has some limitations. First, this study was unblinded and not randomized, and this could have an impact on the results. Thus, as this study had a retrospective design, bias could not be controlled. However, this design is the best to represent routine practice as the attitude of physicians and patients is not influenced by participating in the study. On the other hand, other factors that could have had an impact on the management and clinical end points of the study were not controlled. As a result, the results of the study cannot be exclusively associated with the implementation of the programs. Despite that, the improvements observed in these variables were consistent in the three continuity of care models and with previous studies. Although it would have been interesting to provide information about differences in populations’ information on socioeconomic and health literacy disparities, unfortunately, these data were not available. However, the inclusion criteria were similar in the three hospitals and the three hospitals were part of the Spanish public Health Care system, reducing this potential bias. Finally, the results of our study can only be applied to sanitary areas with similar healthcare system.

Conclusion

Improving continuity of care between primary care and cardiology is associated with improvements in both, management end points (reduction of the delay of first visits to cardiology), and clinical end points (higher intensification of lipid lowering therapy and greater LDL cholesterol control rates). Although the three continuity of care models significantly improved both end points, it seemed that integrating the cardiologist into the primary care center would be more effective than telecardiology in reducing the delay of visits to the cardiologist, and telecardiology together with internal audits and specific medical education programs would be more effective than telecardiology alone or integration of the cardiologist into the primary care center in reducing LDL cholesterol levels. However, further research is required to ascertain whether improving continuity of care is a cost-effective approach.